Title: Anomaly detection in static networks using egonets
Authors: Srijan Sengupta - Virginia Tech (United States) [presenting]
Abstract: Network data has rapidly emerged as an important and active area of statistical methodology. We consider the problem of anomaly detection in networks. Given a large background network, we want to detect whether there is a small anomalous subgraph present in the network, and if such a subgraph is present, we want to identify nodes constitute the subgraph. We propose an inferential tool based on egonets to answer this question. The proposed method is simple and easily extends to several network models, while being computationally efficient and naturally amenable to parallel computing. Using synthetic networks, we demonstrate that the egonet method works well under a wide variety of network models. We obtain interesting results by applying the method on several well-studied benchmark datasets.